- tz - lists.iana.org

Tzcode2006e.tar.gz and tzdata2006e.tar.gz
by Olson, Arthur David (NIH/NCI) [E] May 1, 2006

May 1, 2006

The files... ftp://elsie.nci.nih.gov/pub/tzcode2006e.tar.gz ...and... ftp://elsie.nci.nih.gov/pub/tzdata2006e.tar.gz ...are now available. These reflect the changes circulated on the time zone mailing list last week by Paul Eggert; I've also added new items to tz-arts.htm that were found by Paul. --ado

1 0

RE: Some modifications of China related timezone info.
by Paul Schauble May 1, 2006

May 1, 2006

I agree with the drawbacks of using the most populated city as a proxy for a time zone. In this case, what's wrong with calling the tz zone Asia/China and being done with it? ++PLS -----Original Message----- From: tz-request(a)elsie.nci.nih.gov [mailto:tz-request@elsie.nci.nih.gov] On Behalf Of Zhe Su Sent: Saturday, April 29, 2006 7:06 AM To: Paul Eggert Cc: tz(a)lecserver.nci.nih.gov Subject: Re: Some modifications of China related timezone info. Hi, And in my opinion, using the rule of using most populous location to identify each region is not a generic way for every countries. IMO, we should use different rules for different countries to fit their local circumstances. For China (or more exactly, PRC), it's apparently not suitable to use the population rule. Because the most populous location in China is Chongqing, which is belong to Sichuan province, the most populous province of China. But Chongqing is far less developed than Shanghai and Beijing, so it's far less popular than Beijing and Shanghai. So for China, I'd prefer to use the most important and popular location, Beijing, the capital of China to identify our unique timezone. Regards James Su On 4/29/06, Zhe Su <james.su(a)gmail.com> wrote: > Hi, > I still can't agree with you. It's harmless to add "Asia/Beijing". > And China now uses only one timezone, which is called officially > "China Standard Time" and indeed it's known as Beijing Time. > Now we have five timezones for China in asia/zone.tab: > "Asia/Shanghai", "Asia/Chongqing", "Asia/Urumqi", "Asia/Harbin", > "Asia/Kashgar". But they are known as historic timezones (according to > comments in asia file). After 1949, the establishing of People's > Republic of China, we uses only one timezone, which is called "Beijing > Time" or "China Standard Time" officially. > So if you think we should choose a most populous city of China to > represent the only one "China Standard Time", that is Chongqing rather > than Shanghai, because Chongqing has more than 27 million population. > But, I still think using "Asia/Beijing" to represent the only timezone > used in Peoples Republic of China is the most reasonable way. > > Regards > James Su > > On 4/29/06, Paul Eggert <eggert(a)cs.ucla.edu> wrote: > > "Zhe Su" <james.su(a)gmail.com> writes: > > > > > So, is it ok to just add "Asia/Beijing" into the database along with > > > "Asia/Shanghai"? So that people can choose whatever they want. > > > > For now I'd rather not do that, for the reasons already mentioned. > > >

2 1

Re: Some modifications of China related timezone info.
by Paul Eggert April 29, 2006

April 29, 2006

Ken Pizzini <tz.(a)explicate.org> writes: > the populations do seem to have converged significantly over the 15 > or so years since the Asia/Shanghai entry was created, and the trend > lines strongly suggest that if Beijing has not yet overtaken > Shanghai as the largest city in China, it is very likely to do so > soon. That depends on whose estimates one uses, and what one means by "soon". If we take estimates recently published by Chinese news agencies for the end of 2005, for example, Shanghai had 17.78 million people with a 2.030% annual growth rate since 2000 <http://english.sina.com/china/1/2006/0406/72038.html>, whereas Beijing had 15.36 million people with a 2.135% annual growth rate in the same period <http://news.xinhuanet.com/english/2006-04/20/content_4451085.htm>, so at current rates it will take about 142 years for Beijing's population to surpass Shanghai, at which point both cities will have populations of about 309 million people. To put it mildly, I don't have a lot of faith in these estimates, but until this thing settles out with a clear winner there's no rush. There is an advantage of keeping the Zone names stable, which our other correspondents have reminded us of periodically. I'd rather not get into the business of discussing zone names changes based on guesswork forecasts of population growth. (It's hard enough just keeping track of the clocks. :-)

2 4

Re: Wisconsin's variation of CDT
by Larry M. Smith April 28, 2006

April 28, 2006

Paul Eggert wrote: > Robert Elz <kre(a)munnari.oz.au> writes: > >> Do you really believe that almost anyone (members of this list >> perhaps >> excepted) actually changes their clock at the precise instant that >> DST (or summer time, or whatever it gets called in different areas) >> actually changes. > > Yes, I do. Many people in this country no longer use watches; they > use cell phones to tell time. And cell phones did change at 02:00 in > Wisconsin on April 2; see > <http://www.wisbusiness.com/index.iml?Article=58272>. > <quote> At 2 a.m. on April 2, most of the United States turns clocks ahead one hour as Daylight Saving Time begins. The clock function on wireless phones is controlled by information in the signal from the network on which the phone is located. U.S. Cellular uses Global Positioning Satellite (GPS) technology to set the time on its network, transferring that information to customer handsets. At 2 a.m., the network adjusts the wireless handsetÆs clock for Daylight Saving Time automatically, making each phone an accurate reflection of the current local time. </quote> This argument is starting to recurse into itself; 1) If people in the state of Wisconsin tell time based on electronic equipment. 2) That equipment is keyed off of dataset that is incorrect for them. 3) That dataset should stay the way it is because this is how the people of Wisconsin tell time. Isn't this a circular argument? Please don't get me wrong... I am just publishing my finding, and I expect that before March of 2007 this will all be a moot point. When Wisconsin changes its law (as its dates will be wrong) I expect it will remove date and time and refer back to the U.S. law. So I expect these finding to be of historical nature by the time anyone outside of this list sees them. > I myself prefer an old-fashioned pocket watch, but I've observed that > my students, whose ages are typically around 20, use cell phones or > laptops to tell time just as often as they use watches -- perhaps even > more so. Sales of cheap watches reportedly declined 10% last year in > the US, most likely due to the popularity of telling time by cell > phone; see > <http://www.news-journalonline.com/NewsJournalOnline/News/Headlines/frtHEAD0…>. <quote> "Nowadays with all the electronic goods such as cell phones, iPods, the younger generation is more techno-savvy and they're relying upon electronics that they have and not wearing the watches." </quote> Again, using data based on tzdata. > > I imagine that people who have to log times (nurses, doctors, police > officers, and fire departments, for example) make sure their clocks > are close to the consensus time. These days no doubt a lot of this is > automated (mostly by our work, by Microsoft's, and by the cell phone > companies') but that still counts. > Again, based on tzdata. > I did consult some sources in Wisconsin for what people actually do, > and they agreed that 02:00 was the time people switched. See, for > example, > > Bob Petrie, Locals prepare to lose sleep, gain daylight, Sheboygan > Press (2006-04-01) > <http://www.htrnews.com/apps/pbcs.dll/article?AID=/20060401/SHE0101/60 > 4010508/1062> <quote> At 2 a.m. Sunday, Wisconsin makes the switch, which pushes sunset 60 minutes later into the evening. Sunday's sunset will be at 7:20 p.m. </quote> It can be argued that this reporter hadn't checked his facts. > > >> More likely to be relevant would be what time the radio stations >> announce during the "different" hour > > I'd be surprised if radio-station time announcements in Wisconsin > differed materially from the federal daylight-saving rules. > I have a number of phone calls out to different radio stations in regards to this, but it would not surprise me to find that they are using some device to track time and that it is linked back to using tzdata.

1 0

RE: Some modifications of China related timezone info.
by Andy Lipscomb April 28, 2006

April 28, 2006

One other issue that may or may not be worth considering. Beijing mean time was, prior to the introduction of time zones, the standard for determining midnight in the computation of the Chinese calendar. Thus, having an entry with Beijing's mean time and transition point may be a help to programmers seeking to implement that calendar.

1 0

Re: Some modifications of China related timezone info.
by Zhe Su April 28, 2006

April 28, 2006

Hi, So counting both kind of population, as Beijing and Shanghai have almost same amount of population (though Shanghai has slightly more population right now), and as Beijing is capital of China. I just suggest to add Asia/Beijing into timezone database along with Asia/Shanghai. Just like what was done in my proposed patch. Regards James Su On 4/28/06, Ken Pizzini <tz.(a)explicate.org> wrote: > On Fri, Apr 28, 2006 at 03:52:53PM +0800, Zhe Su wrote: > > I don't think we should count floating population, because those > > part of population is "floating", many of them will move around > > frequently. > > While _individuals_ in this category may come and go, in the aggregate > the "floating" population in these areas has been consistently and > continually growing. They are a real population. > The statistics I've seen (at least the ones that were of good > quality) only reported the *officially registered* portion of these > populations, so clearly the actual population of each urban > agglomeration is actually higher than typically reported, but I'm not > sure how this might affect the relative populations of Beijing and > Shanghai. > > > And the data source that you mentioned is apparently wrong. > > It's not that simple. The data sources I've been able to track down > are all over the map as to the absolute number associated with each > city. Part of this is just that they are reporting based on different > criteria: is the registered transient population included? > Is the unregistered transient population estimated? Is the census > just of the city proper, or does it include the entirety of the "urban > agglomeration"? How recently did the enumeration take place? > And what methodology was used to collect the information (official > census vs. statistical sampling, for example)? > > Except for the fact that it is five years out of date in a region of > very high demographic flux, the document that I would have trusted > most is the Fifth National Census (done in 2000-2001), which still > rated Shanghai more populous by a significant amount. > But clearly that isn't sufficient, so let's compare the value that you > quoted from a Dec. 2005 page on www.bj.xinhuanet.com with a page I > found from Jan. 2005 (in English): > > http://www.chinadaily.com.cn/english/doc/2005-01/10/content_407463.htm > If the two sets of values are presumed to both be correct, then that > means that Beijing's population grew by an astounding 10% over the > course of the year! (Almost all from immigration, since you also > quoted the 0.07% "natural" growth rate.) Surely there must be a > disagreement between how the National Population and Family Commission > of China (quoted by xinhuanet) and the Beijing Municipal Public > Security Bureau (quoted by chinadaily) do their counting. > This is the very kind of discrepancy that makes determining "the" > truth very challenging. > > > > So I suggest the following: the populations do seem to have converged > significantly over the 15 or so years since the Asia/Shanghai entry > was created, and the trend lines strongly suggest that if Beijing has > not yet overtaken Shanghai as the largest city in China, it is very > likely to do so soon. So I say that Asia/Beijing and Asia/Shanghai > should just be preemptively linked: we're talking concentrations of > over 15 million souls in each case anyway, which is surely significant > enough to be worth at least as much as the Africa/Timbuktu link? (As > to which is listed as primary and which as the link, I really don't > care. Toss a coin.) > > --Ken Pizzini >

1 0

Re: Some modifications of China related timezone info.
by Ken Pizzini April 28, 2006

April 28, 2006

On Fri, Apr 28, 2006 at 03:52:53PM +0800, Zhe Su wrote: > I don't think we should count floating population, because those > part of population is "floating", many of them will move around > frequently. While _individuals_ in this category may come and go, in the aggregate the "floating" population in these areas has been consistently and continually growing. They are a real population. The statistics I've seen (at least the ones that were of good quality) only reported the *officially registered* portion of these populations, so clearly the actual population of each urban agglomeration is actually higher than typically reported, but I'm not sure how this might affect the relative populations of Beijing and Shanghai. > And the data source that you mentioned is apparently wrong. It's not that simple. The data sources I've been able to track down are all over the map as to the absolute number associated with each city. Part of this is just that they are reporting based on different criteria: is the registered transient population included? Is the unregistered transient population estimated? Is the census just of the city proper, or does it include the entirety of the "urban agglomeration"? How recently did the enumeration take place? And what methodology was used to collect the information (official census vs. statistical sampling, for example)? Except for the fact that it is five years out of date in a region of very high demographic flux, the document that I would have trusted most is the Fifth National Census (done in 2000-2001), which still rated Shanghai more populous by a significant amount. But clearly that isn't sufficient, so let's compare the value that you quoted from a Dec. 2005 page on www.bj.xinhuanet.com with a page I found from Jan. 2005 (in English): http://www.chinadaily.com.cn/english/doc/2005-01/10/content_407463.htm If the two sets of values are presumed to both be correct, then that means that Beijing's population grew by an astounding 10% over the course of the year! (Almost all from immigration, since you also quoted the 0.07% "natural" growth rate.) Surely there must be a disagreement between how the National Population and Family Commission of China (quoted by xinhuanet) and the Beijing Municipal Public Security Bureau (quoted by chinadaily) do their counting. This is the very kind of discrepancy that makes determining "the" truth very challenging. So I suggest the following: the populations do seem to have converged significantly over the 15 or so years since the Asia/Shanghai entry was created, and the trend lines strongly suggest that if Beijing has not yet overtaken Shanghai as the largest city in China, it is very likely to do so soon. So I say that Asia/Beijing and Asia/Shanghai should just be preemptively linked: we're talking concentrations of over 15 million souls in each case anyway, which is surely significant enough to be worth at least as much as the Africa/Timbuktu link? (As to which is listed as primary and which as the link, I really don't care. Toss a coin.) --Ken Pizzini

1 0

Re: Some modifications of China related timezone info.
by Zhe Su April 28, 2006

April 28, 2006

Hi, I don't think we should count floating population, because those part of population is "floating", many of them will move around frequently. And the data source that you mentioned is apparently wrong. Regards James Su On 4/28/06, Paul Eggert <eggert(a)cs.ucla.edu> wrote: > Robert Elz <kre(a)munnari.oz.au> writes: > > > Having a simple rule that always gets followed allows us to ignore > > endless arguments > > Yes, that's one of the principal motivations. > > > Of course, if it is a close thing, and it is possible that Shanghai > > may overtake Beijing again next year (or something) then keeping > > switching backwards and forwards would be irrational > > Yes, in the past when this has happened, we've tried to avoid change > for the sake of tracking the current ephemeral champion. Of course > this has to be done within reason; if the new champion is undisputed > then it would be time to change. > > But as far as Beijing vs Shanghai goes, it seems that the current > headcount leader is still Shanghai, if we include the floating > population and use the statistics James Su mentioned. > > Also, if we count the entire metropolitan-area population, the > statistics I've seen suggest that Beijing still trails well behind > Shanghai. See <http://www.citypopulation.de/World.html>, for example, > which gives 12.1 million for Beijing and 18.2 million for Shanghai. > Or > <http://www.world-gazetteer.com/wg.php?men=gcis&dat=32&geo=-54&srt=pna > n&col=aohdq&pt=a&srt=pnan>, which gives 11.5 million for Beijing and > 14.9 million for Shanghai. >

1 0

Re: Some modifications of China related timezone info.
by Paul Eggert April 28, 2006

April 28, 2006

Robert Elz <kre(a)munnari.oz.au> writes: > Having a simple rule that always gets followed allows us to ignore > endless arguments Yes, that's one of the principal motivations. > Of course, if it is a close thing, and it is possible that Shanghai > may overtake Beijing again next year (or something) then keeping > switching backwards and forwards would be irrational Yes, in the past when this has happened, we've tried to avoid change for the sake of tracking the current ephemeral champion. Of course this has to be done within reason; if the new champion is undisputed then it would be time to change. But as far as Beijing vs Shanghai goes, it seems that the current headcount leader is still Shanghai, if we include the floating population and use the statistics James Su mentioned. Also, if we count the entire metropolitan-area population, the statistics I've seen suggest that Beijing still trails well behind Shanghai. See <http://www.citypopulation.de/World.html>, for example, which gives 12.1 million for Beijing and 18.2 million for Shanghai. Or <http://www.world-gazetteer.com/wg.php?men=gcis&dat=32&geo=-54&srt=pnan&col=…>, which gives 11.5 million for Beijing and 14.9 million for Shanghai.

1 0

Saskatchewan (2007)
by Chris Walton April 28, 2006

April 28, 2006

Over the past couple of months I have posted a number of e-mails related to Canadian time legislation. I never mentioned Saskatchewan because most of the province does not observe daylight saving time. There are actually two small regions in Saskatchewan that do use daylight saving time: 1)Lloydminster and surrounding area. 2)Creighton and Denare Beach area. Both of these regions are briefly mentioned in TZDATA's "northamerica" file. ################# The city of Lloydminster straddles the Alberta/Saskatchewan border. It uses MST/MDT. There is a special bit of legislation called the "Lloydminster Charter". The bylaws in the charter cover the entire city (both sides of the provincial border). Here are two URLs for the charter. Government of Alberta web site: http://www.qp.gov.ab.ca/Documents/REGS/1979_043.CFM Government of Saskatchewan web site: http://www.qp.gov.sk.ca/documents/english/Statutes/Historical/lloydchr.pdf Section 5(2) reads: ------------------------------------------------------------------ "Whenever any particular time of day is referred to in this Charter, the same shall mean Mountain Standard Time except that in every year during the period between two o'clock in the forenoon of the first Sunday in April and two o'clock in the forenoon of the last Sunday in October it shall mean Central Standard Time." ----------------------------------------------------------------- Since Lloydminster has traditionally followed the daylight savings practices of Alberta, I assume that the charter will have to be updated for 2007. ################# The Battle River region surrounding the city of Lloydminster uses the same time as Lloydminster. Presumably the bylaws in the Lloydminster Charter don't apply outside the city limits. So the region surrounding Lloydminster on the east side of the provincial border is subject to the rules in the Saskatchewan Time Act. The Saskatchewan Time Act defines time for all regions of the province except the city of Lloydminster. The Act allows the rules in western parts of the province to be overridden by public election. http://www.qp.gov.sk.ca/documents/english/statutes/statutes/t14.pdf Several sections of the act are of interest: ---------------------------------------------------------------- 2(i) "summer period means that period in each year beginning at two o'clock, mountain standard time, in the forenoon of the first Sunday in April and ending at two o'clock, mountain standard time, in the forenoon of the last Sunday in October;" --------------------------------------------------------------- "Non-application of Act 5 Nothing in this Act applies to a reference to time in: (a) The Lloydminster Charter; or"... --------------------------------------------------------------- "Time in certain areas of western Saskatchewan 8(1) The time to be used and observed in western Saskatchewan: (a) during the summer period, is central standard time; and (b) during the winter period, is mountain standard time, except in respect of any time option area where central standard time is decided upon by a majority of the electors of the time option area voting with respect thereto under this Act whose ballots are not rejected. (2) Notwithstanding subsection (1), on and from the commencement of the winter period in 1983 and until a vote is taken in accordance with this Act, the time to be used and observed: (a) in the Battle River Time Option Area: (i) during the summer period, is central standard time; and (ii) during the winter period, is mountain standard time; and (b) throughout the year in the Paynton Time Option Area is central standard time." --------------------------------------------------------------------- Notice that the transition from daylight time to standard time is supposed to take place at 2:00am MST (3:00MDT). This is one hour after the transition defined in the Lloydminster charter. Therefore, the area surrounding Lloydminster on the east side of the provincial border should be out of sync with Lloydminster for one hour each year. Ugh! Doesn't this call for a new time zone for a region that probably has no more than 2 or 3 thousand residents. Maybe we just have to ignore the Time Act. The NRC seems to have ignored it. The INMS-NRC site says: "Areas around Lloydminster are in the Mountain Time zone and change at 2:00a.m. local time as in Alberta" http://inms-ienm.nrc-cnrc.gc.ca/en/faq_time_e.php#5 ################ And finally there is the small area that contains the towns of Creighton and Denare Beach. This region uses CST/CDT to align with the city of Flin Flon Manitoba. The use of daylight saving time in this region is in defiance of the Saskatchewan Time Act. Should we stop and ponder what time Creighton and Denare Beach have historically done the transition from daylight time to standard time? Is it: a) 2:00am wall time as is tradition in most parts of the country or b) 02:00 CST (3:00am wall time) as per the old Manitoba Time Act or c) 02:00 MST (4:00am wall time) as per the definition of "summer time" in the Saskatchewan Time Act? or d) 8:07am wall time because we can't expect non-law-abiding citizens get up in the middle of the night to change the clocks. It will be interesting to see if the Saskatchewan government bothers to update the Time Act to reflect the extra four weeks of daylight savings in 2007. Personally I think that the all fourteen pages of the old act should be burned and replaced with something more concise. But that is just my opinion. I will keep an eye out for new legislation and I will post an e-mail to this group if I see anything. -chris

1 0